Registry / llm-agents / anyscale

anyscale

JSON →
library0.26.108pypypi✓ verified 23d ago

Anyscale is a Python library and CLI that provides programmatic tools for authenticating and interacting with the Anyscale platform. It simplifies the deployment and scaling of machine learning models, particularly those involving large language models (LLMs), by enabling interaction with Anyscale's cloud infrastructure from Python applications or the command line. The library is actively maintained, with frequent updates, and the current version is 0.26.93.

pip install anyscale
INSTALL
IMPORT
SIG · ANYSCALE
A
anyscale
llm-agentspythonv0.26.108
Install
20.1s avg
Import
4443ms
Disk
439MB
Pass rate
10/ 10
Env Coverage10 / 10
glibc
3.93.13
musl
3.93.13
Install & Compatibility
Where this runs
tested against v0.26.108 · pip install
no network on importno background threads
Install × environment matrix
Each cell = how many times install + import succeeded across repeated harness runs. Partial = flaky.
glibc = Debian/Ubuntu slim · musl = Alpine Linux
musl
py 3.103.915 runs
installs and imports cleanly · install 0.0s · import 4.649s · 433.5MB
glibc
py 3.103.915 runs
installs and imports cleanly · install 20.1s · import 4.237s · 434MB
439MB installed
● package 439MB
Code
Verified usage

Verified import paths — ran on the pinned version, not inferred.

Anyscale
from anyscale import Anyscale
import anyscale
AnyscaleClient
from anyscale import AnyscaleClient
import anyscale
JobSDK
from anyscale import JobSDK
import anyscale

This quickstart demonstrates how to programmatically list Anyscale jobs using the SDK. Before running this code, you must authenticate the Anyscale CLI by running `anyscale login` in your terminal or by setting the `ANYSCALE_CLI_TOKEN` environment variable for non-interactive environments.

import os import anyscale # Ensure you have logged in via `anyscale login` or set ANYSCALE_CLI_TOKEN # For non-interactive use, generate an API key from the Anyscale console # and set it as an environment variable. # os.environ['ANYSCALE_CLI_TOKEN'] = os.environ.get('ANYSCALE_CLI_TOKEN', '') # List Anyscale jobs try: print("Listing Anyscale jobs...") jobs = anyscale.job.list(max_items=5) if jobs.results: for job in jobs.results: print(f" Job ID: {job.id}, Name: {job.name}, State: {job.state}") else: print(" No Anyscale jobs found.") except Exception as e: print(f"Error listing jobs: {e}") print("Please ensure you are authenticated. Run `anyscale login` or set the ANYSCALE_CLI_TOKEN environment variable.")
anyscale --version
Debug
Known issues
breakingThe legacy `AnyscaleSDK` class is deprecated and will be removed by February 28, 2026. Migrate to directly importing `anyscale` and using `anyscale.<module>.<function>()`.
fix
Replace `from anyscale.sdk import AnyscaleSDK; sdk = AnyscaleSDK()` with `import anyscale` and call methods like `anyscale.job.list()`.
affects: 0.26.86 and earlier
breakingLegacy CLI commands like `anyscale cluster-env`, `anyscale cluster start/terminate`, and `anyscale machine list/delete` have been removed. Use `anyscale image` and `anyscale workspace_v2` commands instead.
fix
Consult the Anyscale CLI documentation for the equivalent new commands, e.g., use `anyscale image` for image management and `anyscale workspace_v2` for workspace operations.
affects: 0.26.77 and later
breakingSupport for Python 3.8 was removed starting in Anyscale CLI version 0.26.75.
fix
Upgrade your Python environment to version 3.9 or higher (up to <3.14).
affects: 0.26.75 and later
deprecatedThe `anyscale cloud edit` command is deprecated. Use `anyscale cloud update` instead.
fix
Replace calls to `anyscale cloud edit` with `anyscale cloud update`.
affects: 0.26.81 and later
deprecatedLegacy cloud SDK methods `get_cloud`, `get_default_cloud`, and `search_clouds` have been removed. Use `anyscale.cloud.get`, `anyscale.cloud.get_default`, and `anyscale.cloud.list` respectively.
fix
Update SDK code to use the new `anyscale.cloud` module methods.
affects: 0.26.85 and later
gotchaThe default value for `max_retries` when submitting jobs will change from 1 to 0 in a future release. Explicitly set `max_retries` for consistent behavior.
fix
Always explicitly set the `max_retries` parameter when submitting Anyscale jobs to avoid unexpected behavior changes in future releases.
affects: All versions, proactive change recommended
Upgrade
Version history
0.26.108latest on PyPI · released Aug 18, 2026
Audit
Dependencies
pythonrequiredRuntime environment
Agent activity
66 hits · last 30 days
node
56
OpenAI (training)
1
Resources
anyscale — pip install anyscale · libregistry